Xerith

Xerith is a systems-oriented programming language designed for granular control over the compilation pipeline. This implementation serves as a technical exploration of recursive descent parsing, Abstract Syntax Tree (AST) architecture, and tree-walk evaluation.

Architecture

The project is structured as a standard compiler front-end:

• Lexer: A hand-written scanner for deterministic tokenization.
• Parser: A recursive descent implementation with operator precedence handling.
• AST Implementation: A strongly-typed tree structure for intermediate representation.
• Interpreter: A visitor-pattern based evaluator that decouples execution logic from node definitions.

Key Design Principles

• Zero Dependencies: The core implementation avoids external libraries to ensure a minimal, build-stable binary.
• AST Integrity: Prioritizing strict node definitions to facilitate easier semantic analysis in future iterations.
• Transparent Tooling: Integrated CLI for debugging token streams and AST dumps.

Implementation Roadmap

• Lexical analysis and Token stream generation
• Abstract Syntax Tree (AST) core nodes
• Visitor Pattern infrastructure for evaluation
• Environment management and Lexical Scoping
• Control flow primitives (If/While)
• Function declarations and stack frame handling
• Bytecode IR and Virtual Machine (Target)

Build System

Xerith uses CMake for build orchestration and is developed primarily in a C++17/20 environment.

mkdir build && cd build
cmake ..
make
./xerith path/to/script.xrtx

Trademark & Licensing

The name “Xerith” is a registered trademark of NerdBlud. The source code is licensed under AGPL v3. You are free to fork and modify the implementation, but redistribution under the "Xerith" name or confusingly similar branding requires explicit permission.

---

Developed by nerdblud